摘要:
Direct compression of original map can’t meet users' request on image quality. In order to harmonize image quality and bandwidth of wireless transmission, a protect scheme of UEP (Unequal Error Protection) with JEPG 2000 coder and Channel coding based on RCPC (Rate Compatible Punctured Convolutional) was proposed to improve ROI-based image quality. Firstly, ROI (region-of-interest) was extracted from Saliency Map based on visual attention model. Secondly, ROI and BG (background region) in image which transmitted in wireless channel respectively, were protected by different encoding bit rates and RCPC Code rates. Finally, the PSNR of image was calculated after transmitting in wireless channel using computer simulations. Experimental results show that ROI-based images with high-level protection get a higher quality than image without ROI while not increasing bandwidth.
期刊:
Proceedings of SPIE - The International Society for Optical Engineering,2012年8349 ISSN:0277-786X
通讯作者:
Zhao, G.(zhaogang@mail.ccnu.edu.cn)
作者机构:
[Zhao, Gang] National Engineering Research Center for E-Learning, Huazhong Normal University, Wuhan, China;[Li, Yangyang; Zhao, Gang; He, Bin; Liu, Qifang] Department of Information Technology, Huazhong Normal University, Wuhan, China
会议名称:
4th International Conference on Machine Vision (ICMV) - Machine Vision, Image Processing and Pattern Analysis
会议时间:
DEC 09-10, 2011
会议地点:
Singapore, SINGAPORE
会议论文集名称:
Proceedings of SPIE
关键词:
video stitching;camera pose estimation;boundary resampling
摘要:
This paper concerns the problem of video stitching automatically in a multi-camera surveillance system. Previous approaches have used multiple calibrated cameras for video mosaic in large scale monitoring application. In this work, we formulate video stitching as a multi-image registration and blending problem, and not all cameras are needed to be calibrated except a few selected master cameras. SURF is used to find matched pairs of image key points from different cameras, and then camera pose is estimated and refined. Homography matrix is employed to calculate overlapping pixels and finally implement boundary resample algorithm to blend images. The result of simulation demonstrates the efficiency of our method.
期刊:
Proceedings of SPIE - The International Society for Optical Engineering,2012年8350:587-591 ISSN:0277-786X
通讯作者:
Hu, Yan
作者机构:
[He, Bin; Yang, Yongchao; Hu, Yan; Zhao, Gang; Shen, Guosheng] Huazhong Normal Univ, Natl Engn Res Ctr E Learning, Wuhan 430079, Hubei, Peoples R China.
通讯机构:
[Hu, Yan] H;Huazhong Normal Univ, Natl Engn Res Ctr E Learning, Wuhan 430079, Hubei, Peoples R China.
会议名称:
2010 International Conference on Software and Computing Technology(2010年软件与计算机技术国际会议 ICSCT 2010)
会议时间:
2010-10-17
会议地点:
昆明
会议主办单位:
[Hu, Yan;Yang, Yongchao;Zhao, Gang;He, Bin;Shen, Guosheng] Huazhong Normal Univ, Natl Engn Res Ctr E Learning, Wuhan 430079, Hubei, Peoples R China.
会议论文集名称:
2010 International Conference on Software and Computing Technology(2010年软件与计算机技术国际会议 ICSCT 2010)论文集
关键词:
Model Transformation;Large-scale three-Dimensional Scene model;Model Optimization;Data loss;Realtime rendering
摘要:
During the simulation process of real-time three-dimensional scene, the popular modeling software and the real-time rendering platform are not compatible. The common solution is to create three-dimensional scene model by using modeling software and then transform the format supported by rendering platform. This paper takes digital campus scene simulation as an example, analyzes and solves the problems of surface loss; texture distortion and loss; model flicker and so on during the transformation from 3Ds Max to MultiGen Creator. Besides, it proposes the optimization strategy of model which is transformed. The operation results show that this strategy is a good solution to all kinds of problems existing in transformation and it can speed up the rendering speed of the model.
摘要:
In order to improve the educational function of e-learning, a virtual classroom with intelligent virtual tutor has been developed. Architecture of the virtual classroom system is presented. The intelligent virtual tutor can perceive environment and make its own decision to enhance the interaction between tutor and remote students in some degree. The virtual tutor is emotive, which can help remote students enhance their understanding of the content. The system presented is used to learn course "Electronics" for sophomore. It's for blended education, as a necessary supplement for real-life and regular classroom teaching and learning, which is competent to achieve tutoring tasks.
作者机构:
[Gang, Zhao; Ya, Liusan; Ling, Zhao] Engineering and Research Center for Information Technology on Education, CCNU, Wuhan 430079, China
通讯机构:
Engineering and Research Center for Information Technology on Education, CCNU, China
关键词:
Digital Education;Resources Supermarket;Service Pattern
摘要:
Recently, Digital Education has been faced with two primary problems: messy and low-sharing. In order to resolve those problems, the paper suggests a new Service Pattern which combines the traditional educational resources service and E-commerce service, analyses relationships and realizations between Service Factors and Service Strategies in the abstract, designs a Service Model based on the Service Pattern, and bring forward a corresponding mechanism of resources management. Finally, in order to prove its technical feasibility, the paper realizes the Service Pattern by a Digital Educational Resources Supermarket system which is based on Web 2.0 and J2EE.
期刊:
2008 IEEE INTERNATIONAL CONFERENCE ON GRANULAR COMPUTING, VOLS 1 AND 2,2008年:121-124
通讯作者:
Zhao Gang
作者机构:
[Zhao Gang; Chen Di] Huazhong Normal Univ, Engn Res Ctr Educ Informat Technol, Wuhan, Peoples R China.;[Zhao Gang] Huazhong Normal Univ, Dept Informat Technol, Wuhan, Peoples R China.;[Xu Juhong] Wuhan Inst Technol, Art Design, Wuhan, Peoples R China.
通讯机构:
[Zhao Gang] H;Huazhong Normal Univ, Dept Informat Technol, Wuhan, Peoples R China.
关键词:
virtual smart classroom;distance education;design plan
摘要:
With the development of the network and computer technology, the educational technology based on computer network is becoming popular worldwide. However, nowadayspsila distance education merely stays on the stage of Web-based courseware and one-way teaching, and it can not effectively support real-time classroom teacher-student interaction. This paper proposes a new interactive distance teaching and demonstration mode, which can allow the interaction between the teacher and the virtual scene as well as between the teacher and the student. We emphasize the design of the smart classroom that based on blending reality. In this paper, It shows the train of thoughts that how to integrate the virtual reality technology into distance teaching to help teachers show their teaching materials. We also present the method of that teachers interact on the object in the virtual reality.
摘要:
With the continuing growth of the internet, more and more people arc taking advantage of distance learning courses. However, most of distance education systems have not been able to prefectly consider distinct personality differences between learners. Learners can only accept identical learning contents, but can not choose their learning processes and environments freely according to their own interests, study styles and demands. In addition, some existing systems have some difficulties to organize learning resources into an integrated course, with regardless of personalized learning environment. To solve these problems, this paper proposes a personalized distance education system model. The system is based on web mining technology and Moodle platform. Through using web mining, the system can discover paths accessed frequently, association rules mode and sequence mode, in order to automatically adjust learners' learning programs and progresses. At the same time, utilizing Moodle platform, it can organize those mined information into an integrated course, and establish personalized learning environment to meet students' personalized demands.
期刊:
KAM: 2008 INTERNATIONAL SYMPOSIUM ON KNOWLEDGE ACQUISITION AND MODELING, PROCEEDINGS,2008年:520-524
通讯作者:
Zhao, Gang
作者机构:
[Liu, Zhiying; Zhao, Gang] Huazhong Normal Univ, Engn & Res Ctr Informat Technol Educ, Wuhan, Peoples R China.;[Liu, Zhiying; Zheng, Weibing; Zhao, Gang] Huazhong Normal Univ, Dept Informat Technol, Wuhan, Peoples R China.;[Jin, Jesse] Newcastle Univ, Sch Design, Commun & Informat Technol, Newcastle, NSW, Australia.
通讯机构:
[Zhao, Gang] H;Huazhong Normal Univ, Engn & Res Ctr Informat Technol Educ, Wuhan, Peoples R China.
摘要:
Mobile-learning (in-learning) is not a new word in education nowadays, since it can be carried out time and space independently. However, it is difficult to be used for systematic study. In the other hand, web2.0 has been prevailed in internet. The most outstanding traits of web2.0 are harnessing collective intelligence, centering on individual and self-organization, which can increase learners' initiative and enthusiasm. Hence, we propose an idea of integrating web2.0 into in-learning, which will be more efficient to attract learners' attention for periodical m-learning, and knowledge can be shared and recreated better. According to this idea, a m-learning application platform and a system framework based on web2.0 is presented in this paper.